Journal of Microbiological Methods

Page created by Carmen Henry
 
CONTINUE READING
Journal of Microbiological Methods
Journal of Microbiological Methods 185 (2021) 106224

                                                                     Contents lists available at ScienceDirect

                                                      Journal of Microbiological Methods
                                                         journal homepage: www.elsevier.com/locate/jmicmeth

Fast and automated detection of common carbapenemase genes using
multiplex real-time PCR on the BD MAX™ system
Katja Probst a, *, Sébastien Boutin a, Michael Bandilla c, Klaus Heeg a, Alexander H. Dalpke a, b
a
  Department of Infectious Diseases, Medical Microbiology and Hygiene, University Hospital Heidelberg, Heidelberg, Germany
b
  Institute of Medical Microbiology and Virology, Medical Faculty, Technische Universität Dresden, Dresden, Germany
c
  FRIZ Biochem GmbH, München, Neuried, Germany

A R T I C L E I N F O                                      A B S T R A C T

Keywords:                                                  Fast detection of carbapenemases in Gram-negative bacilli is necessary for accurate antibiotic treatment, pre­
BD MAX™                                                    vention of further spreading and surveillance purposes. We analyzed the current occurrence of gene variants and
Carbapenemase                                              designed two multiplex PCRs with hydrolysis probes.
Molecular diagnostics
                                                              The assay was developed for the BD MAX™ system that combines DNA extraction and PCR in a fully auto­
Multidrug-resistant gram-negative
Multiplex PCR
                                                           mated procedure providing results within 3 h and was evaluated for detection of carbapenemases from bacterial
                                                           isolates and directly from rectal swabs.
                                                              The assay has a theoretic coverage of 97.1% for carbapenemases detected during the last years by the German
                                                           National Reference Laboratory (NRL). A collection of 151 isolates from the NRL was used and all carbapenemase-
                                                           positive bacteria (58/58) were identified correctly. The direct-PCR on rectal swabs revealed additional carba­
                                                           penemase genes in 7 samples that were not identified by the culture-based method used as reference method.
                                                              The assay allows detection of carbapenemases from clinical isolates and might also help in rapid detection
                                                           directly from rectal samples.

1. Introduction                                                                                  “the big 4”, are also the most prevalent carbapenemases found in
                                                                                                 Enterobacteriaceae over the last years in Germany (KPC: 14.7%–26.9%,
    Resistance to beta-lactam antibiotics in Gram-negative bacilli is an                         NDM: 18.0%–20.5% VIM: 18.2%–24.4% and OXA-48-like: 31.4%–
increasing problem worldwide. In case of resistance to carbapenems, the                          47.4%) (Koch-Institut, 2015; Koch-Institut, 2016; Pfennigwerth, 2017;
remaining treatment options for severe infections are limited (Kaase,                            Pfennigwerth, 2018). In the non-fermenters Pseudomonas aeruginosa and
2012). Carbapenem resistance occurs mostly via carbapenemases, ac­                               Acinetobacter baumannii, respectively VIM (79.2%–88.2%), IMP (6.5%–
quired enzymes that are able to hydrolyze different beta-lactam anti­                            10.3%) and OXA-23 (66.7%–80.7%), OXA-72 (11.0%–12.7%) and OXA-
biotics. A combination of porin loss or overexpression of efflux pumps                           58 (1.3%–7.0%), were detected more frequently (Koch-Institut, 2015;
with expression of extended-spectrum beta lactamases (ESBLs) or                                  Koch-Institut, 2016; Pfennigwerth, 2017; Pfennigwerth, 2018).
AmpCs, depending on the bacteria, is another way of obtaining                                        Carbapenemases are classified based on their amino acid sequence
carbapenem-resistance (El Amin et al., 2005; Doumith et al., 2009).                              into different Ambler classes. KPC, a serine beta-lactamase, belongs to
Carbapenemase genes are usually located on plasmids that allow facil­                            Ambler class A; NDM, VIM and IMP, also known as metallo-beta-
itated spreading (Queenan and Bush, 2007). The first carbapenemase                               lactamases belong to class B. The oxacillinases OXA-48, OXA-72, OXA-
(NmcA) was discovered in an Enterobacteriaceae in 1993 (Naas and                                 23 and OXA-58 classify in class D (Queenan and Bush, 2007).
Nordmann, 1994). Over the last years, a large variety of carbapenemases                          Numerous variants exist within each class due to mutations of the genes.
have been described but the most important remain Klebsiella pneumo­                             Newly occurring variants are a problem for molecular-based detection
niae carbapenemase (KPC), New-Delhi-metallo-beta-lactamase (NDM),                                methods, thus regular verification of primer coverage is necessary.
Verona Integron-encoded metallo-beta-lactamase (VIM) and the oxa­                                    For infection control it is important to know the mechanism of car­
cillinases (OXA-48) (Albiger, Glasner et al., 2015). These four enzymes,                         bapenem resistance, thus detection methods that cover different

 * Corresponding author at: Department of Infectious Diseases, Medical Microbiology and Hygiene, University Hospital Heidelberg, Im Neuenheimer Feld 324,
69120 Heidelberg, Germany.
   E-mail address: katja.probst@med.uni-heidelberg.de (K. Probst).

https://doi.org/10.1016/j.mimet.2021.106224
Received 18 January 2021; Received in revised form 14 April 2021; Accepted 14 April 2021
Available online 16 April 2021
0167-7012/© 2021 Elsevier B.V. All rights reserved.
Journal of Microbiological Methods
K. Probst et al.                                                                                                      Journal of Microbiological Methods 185 (2021) 106224

carbapenemases and their variants are needed. Besides phenotypic                      isolates harboring resistance genes, not included in our assay (blaGES: n
detection methods of carbapenem resistance, a number of different                     = 5, blaCMY: n = 5, blaSHV: n = 1, blaDHA: n = 1) and beta-lactamase-
genotypic assays based on monoplex or multiplex quantitative real-time                negative bacteria (n = 81). The second set was a random collection
PCR and using hydrolysis probes or melt-curve analysis have been                      with isolates harboring less common carbapenemase variants, especially
developed. Eazyplex® SuperBug CRE (Amplex Biosystems GmbH, Gie­                       different blaIMP and blaOXA variants, and bacteria harboring two
ßen, Germany) or Check-Direct CPE assay (Check-Points, Wageningen,                    different carbapenemases simultaneously (n = 32). Bacterial species
The Netherlands), for example, are designed to detect the “big 4” car­                were determined by MALDI TOF (Bruker Daltonics, Germany). The
bapenemases. There are some assays that include detection of IMP                      isolates were well characterized by several phenotypic tests, like sus­
variants, such as the Cepheid Xpert Carba-R and different ESBLs and                   ceptibility testing, the modified Hodge Test and the combined disk test
AmpCs (Dallenne et al., 2010; Naas et al., 2011; Cuzon et al., 2012). Yet,            with EDTA and boronic acid, additionally PCR and sequencing was
most of the commercially available assays are limited to the “big 4”                  performed at the NRL (Hofko et al., 2014).
carbapenemases commonly found in Enterobacteriaceae and do not                            In addition, bacterial isolates and the corresponding rectal swabs
include all gene variants.                                                            with a known carbapenemase were collected from the University Hos­
    Detection of the numerous variants of carbapenemases by PCR in a                  pital Heidelberg during a procedure of routine patient screening at
routine workflow can be highly facilitated by fully automated solutions,              hospital admission. The isolates were regrown on BD™ Columbia Agar
which can be modulated to answer changes in the local epidemiology.                   with 5% Sheep Blood (Becton Dickinson GmbH, Heidelberg, Germany)
The BD MAX™ system is a fully integrated molecular solution that, as a                at 37 ◦ C.
peculiarity, can be used in an open-system mode, which allows user-
developed protocols. The Check-Direct CPE assay has been used on a                    2.3. DNA extraction
BD MAX™ for detection of carbapenemases (Antonelli et al., 2016) and
additional real-time PCRs covering blaIMP, blaVIM, blaOXA-23 and mcr                     The chelex-based DNA extraction was performed as described by
colistin resistance have been published (Teo et al., 2016; Mentasti et al.,           Martin-Platero et al. in 2010 (Martin-Platero et al., 2010). For purifying
2019). Here, we performed an in-depth analysis of recently occurring                  the DNA ethanol precipitation was carried out, i.e., adding 0.3 M
carbapenemase genes to define new primers for the detection of carba­                 ammonium acetate and 80% (v/v) of ethanol absolute. Samples were
penemases in Enterobacteriaceae and in the non-fermenters P. aeruginosa               inverted and incubated at − 80 ◦ C for one hour. After incubation, the
and A. baumannii.                                                                     samples were centrifuged (20 min, 13,300 rpm, 4 ◦ C). The supernatant
                                                                                      was discarded and two washing steps followed by adding 300 μl of 70%
2. Materials and methods                                                              ethanol, mixing and centrifugation (20 min, 13,300 rpm, 4 ◦ C). After
                                                                                      drying the pellet, it was resuspended with 56 ◦ C nucleic acid-free water.
2.1. Assay design                                                                     DNA quality was determined with the NanoDrop 1000 Spectropho­
                                                                                      tometer (Thermo Fisher Scientific); quantification was done with the
     A sequence collection of beta lactamase genes from the Center for                Quant-iT™ PicoGreen™ dsDNA Assay Kit (Thermo Fisher Scientific).
Genomic Epidemiology (https://cge.cbs.dtu.dk/services/data.php,                          DNA extracted with the chelex-based method was used to analyze the
December 2017) was used to design primers and hydrolysis probes. It                   PCR performance, since a known amount of DNA is needed for deter­
was complemented with carbapenemase gene variants from the NCBI                       mining the limit of detection (LOD). The validation of carbapenemase
database (December 2017, Table S1). The assay was designed to detect                  detection in clinical samples was performed on the BD MAX™ system,
eight common carbapenemase groups: blaNDM, blaKPC, blaVIM, blaIMP,                    by using the BD MAX™ ExK™ DNA-2 Kit for DNA extraction and PCR in
blaOXA-23-like, blaOXA-40/24-like, blaOXA-58-like and blaOXA-48-like.                 an automated closed system.
     The alignment of the carbapenemase sequences was done with
MUSCLE in MEGA 7.0.26. For alignment the following carbapenemase                      2.4. Multiplex PCR in PCR-only mode
variants were selected: blaNDM-1 to blaNDM-21, blaKPC-1 to blaKPC-32, blaVIM-
1 to blaVIM-45, blaIMP-1 to blaIMP-69 and blaOXA-1 to blaOXA-566. The aim was             Quantitative real-time PCRs were performed on the BD MAX™ sys­
to cover as many gene variants as possible with one set of primers and                tem. For detection of all eight carbapenemases two master mixes (MM1
probe. To analyze variants clustering together, phylogenetic trees were               and MM2) with a final volume of 12.5 μl were prepared. Each master
created with the UPGMA method in MEGA 7.0.26. Either all sequences                    mix consisted of 2.5 μl primaQUANT 5× qPCR-Probe-MasterMix
or a selection of sequences were used to create a consensus sequence.                 (Steinbrenner GmbH, Wiesenbach, Germany), a forward (F) and
The design of the primers and probes was done with the IDT primer tool                reverse (R) primer, the hydrolysis probe (P), PCR-grade water and 4 μl
(Integrated DNA Technologies, Inc.). To define two multiplex PCRs,                    DNA. The following final concentrations of primers and probes were
possible interactions between all oligonucleotides were analyzed with                 used for MM1: IMP-F/R, 16S-F/R: 0.3 μM; KPC-F/R, NDM-F/R, VIM-F/
the multiple primer analyzer (Thermos Fisher Scientific). As a positive               R: 0.4 μM; IMP-P (CY®5: 650 nm–670 nm): 0.2 μM, 16S-P (Cy5.5: 684
control primers and probe for the detection of 16S rDNA were included                 nm–708 nm): 0.25 μM, KPC-P (JOE: 529 nm–555 nm), NDM-P (FAM:
to estimate the impact of PCR inhibition on negative results (Nadkarni                495 nm–520 nm), VIM-P (Texas Red®: 597 nm–616 nm): 0.15 μM; and
et al., 2002). The first multiplex PCR was designed to detect blaNDM,                 for MM2: primers: 0.4 μM, OXA-23-P (FAM: 495 nm–520 nm), OXA-40/
blaKPC, blaVIM, blaIMP and 16S rDNA, the second multiplex PCR is for                  24-P (JOE: 529 nm–555 nm): 0.25 μM; OXA-58-P (Texas Red®: 597
detection of blaOXA-23-like, blaOXA-40/− 24-like, blaOXA-58-like and blaOXA-48-       nm–616 nm), OXA-48-P (CY®5: 650 nm–670 nm): 0.15 μM. Then, 10 μl
like. Sequences, amplicon size and melting temperature are detailed in                of MM1 and MM2 were transferred to the BD MAX™ PCR cartridge. PCR
the table S2.                                                                         was run in the PCR-only mode with the following program: 98 ◦ C, 3 min;
                                                                                      98 ◦ C, 5 s; 57 ◦ C, 47.8 s for 3 cycles; 98 ◦ C, 5 s; 61 ◦ C, 43 s for 37 cycles.
2.2. Bacterial isolates
                                                                                      2.5. Limit of detection
    To test and validate the assay, two collections of bacterial isolates
from the German National Reference Laboratory (NRL) were used. Iso­                       To determine the LOD, one representative isolate harboring a
lates of the first collection (n = 151) were consecutively collected during           carbapenemase gene of each group was selected (A. baumannii: blaNDM-
a half-month period (Jan, 21st to Feb, 6th 2013), consisting of                       1/− 6, K. oxytoca: blaKPC-3, E. cloacae: blaVIM-1, P. aeruginosa: blaIMP-8/− 24,
carbapenemase-positive bacteria (blaNDM: n = 5, blaKPC: n = 8, blaVIM: n              A. baumannii: blaOXA-23, A. baumannii: blaOXA-72, A. baumannii: blaOXA-58,
= 23, blaOXA-23: n = 12, blaOXA-48: n = 9, blaNDM and blaOXA-48: n = 1),              S. marcescens: blaOXA-48). Genomic copy numbers were calculated based

                                                                                  2
Journal of Microbiological Methods
K. Probst et al.                                                                                                   Journal of Microbiological Methods 185 (2021) 106224

on the measured DNA concentration and the average genome size of                 Table 1
each bacterial species. Copy numbers were used instead of CFU and DNA            In silico and experimentally detectable carbapenemase variants by multiplex
concentrations, to make interpretation independent on the efficiency of          real-time PCR.
the DNA extraction and the bacterial species. Yet, there can still be             Carbapenemase       In silico detectable variants      Experimentally tested variants
variations due to different copy numbers of the carbapenemase-                    gene
encoding plasmids. As the calculation was done for one copy of the                blaNDM              blaNDM-1-blaNDM-21                 blaNDM-1, blaNDM-5, blaNDM-6
carbapenemase gene, the LOD should be seen as an approximation. The               blaKPC              blaKPC-1-blaKPC-32                 blaKPC-2, blaKPC-3
DNA samples were adjusted to 105 copies/μl and 10-fold serial dilutions           blaVIM              blaVIM-1-blaVIM-45, except         blaVIM-1, blaVIM-2, blaVIM-26,
                                                                                                      blaVIM-7                           blaVIM-28
to one copy/μl were prepared. Based on the slope of the standard curve,
                                                                                  blaIMP              blaIMP-1-3, blaIMP-5-10, blaIMP-   blaIMP-1, blaIMP-2, blaIMP-7,
PCR efficiency was calculated by using the equation: E =
 (             )                                                                                      14-16, blaIMP-18-20, blaIMP-22-    blaIMP-8, blaIMP-13, blaIMP-14,
           − 1
   − 1 +10slope *100. The multiplex PCRs were performed on the BD                                     25, blaIMP-28-30, blaIMP-32,       blaIMP-15, blaIMP-16, blaIMP-19,
                                                                                                      blaIMP-40-43, blaIMP-45,           blaIMP-24, blaIMP-28
MAX™ system in the PCR-only mode described above.                                                     blaIMP-48, blaIMP-49, blaIMP-
                                                                                                      52, blaIMP-53, blaIMP-55,
                                                                                                      blaIMP-56, blaIMP-58, blaIMP-
2.6. Combined extraction and PCR on the BD MAX™ system
                                                                                                      60, blaIMP-62, blaIMP-
                                                                                                      66,blaIMP-69-73, blaIMP-75-80
   The sample buffer tube was inoculated with one bacterial colony.               blaOXA-23-like      blaOXA-23, blaOXA-27, blaOXA-      blaOXA-23
The DNA was extracted with the BD MAX™ ExK™ DNA-2 Kit (4 snap                                         49, blaOXA-73, blaOXA-102,

configuration). To dilute the elution buffer, 50 μl of PCR-grade water                                blaOXA-103, blaOXA-105,
                                                                                                      blaOXA-133, blaOXA-134,
were added to the extraction stripes. For extraction, the BD MAX™                                     blaOXA-146, blaOXA-165-171,
ExK™ DNA-2 protocol (type 3: liquid MM with primers and probes) was                                   blaOXA-225, blaOXA-239,
used. The PCR was performed with the program described above.                                         blaOXA-366, blaOXA-398,
                                                                                                      blaOXA-422, blaOXA-423,
                                                                                                      blaOXA-435, blaOXA-440,
2.7. Carbapenemase detection directly from rectal swabs                                               blaOXA-469, blaOXA-481-483,
                                                                                                      blaOXA-565, blaOXA-657,
    In total, 66 rectal swabs (ESwab™, Copan) with multidrug-resistant                                blaOXA-806-818
Gram-negative bacteria previously detected by selective culture                   blaOXA-40/24-like   blaOXA-40/24, blaOXA-25,           blaOXA-72
                                                                                                      blaOXA-26, blaOXA-72, blaOXA-
screening were collected from the University Hospital Heidelberg in the
                                                                                                      139, blaOXA-160, blaOXA-207,
years 2017–2019. Swabs were selected for direct PCR if bacterial growth                               blaOXA-437, blaOXA-653
on a chromID® ESBL plate, inoculated with 10 μl rectal swab medium,               blaOXA-58-like      blaOXA-58, blaOXA-96, blaOXA-      blaOXA-58
could be observed (bioMérieux Deutschland GmbH). ChromID® ESBL                                       97, blaOXA-164, blaOXA-397,

plates were used to distinguish between the German 3MRGN and                                          blaOXA-420, blaOXA-467,
                                                                                                      blaOXA-512
4MRGN classification. The colony counts on the chromID® ESBL plates               blaOXA-48-like      blaOXA-48, blaOXA-162, blaOXA-     blaOXA-48, blaOXA-162, blaOXA-181,
were used to estimate the bacterial load (CFU/ml) per swab. One or                                    163, blaOXA-181, blaOXA-199,       blaOXA-204, blaOXA-232, blaOXA-244
more representative isolates were selected for analyzing the resistance                               blaOXA-204, blaOXA-232,
to meropenem and imipenem using VITEK®2 system (bioMérieux                                           blaOXA-244, blaOXA-245,
                                                                                                      blaOXA-247, blaOXA-370,
Deutschland GmbH, Nürtingen, Germany). The bacterial species was
                                                                                                      blaOXA-405, blaOXA-416,
determined by MALDI TOF (Bruker Daltonics, Germany). To identify the                                  blaOXA-438, blaOXA-439,
carbapenemase type, isolates were tested by multiplex SYBR green real-                                blaOXA-484, blaOXA-505,
time PCR (Hofko et al., 2014), used as a reference method. Direct PCR                                 blaOXA-514, blaOXA-517,
was performed by adding 100 μl of the rectal swab medium to the                                       blaOXA-519, blaOXA-538,
                                                                                                      blaOXA-547, blaOXA-566,
sample buffer tube of the ExK DNA-2 Kit. Details on the PCR procedure                                 blaOXA-567, blaOXA-788,
are given above.                                                                                      blaOXA-793

3. Results
                                                                                 with serial dilutions in four replicates. For blaNDM, blaVIM, blaOXA-23,
3.1. Primer design                                                               blaOXA-72, blaOXA-58 and blaOXA-48 the detection of 3.2 copies/PCR was
                                                                                 possible. The LOD for blaKPC and blaIMP was 32 copies/PCR. The PCR
   Nine sets of primers and probes were designed to detect eight com­            efficiency for blaNDM, blaOXA-58, blaOXA-72 and blaOXA-48 was in the
mon carbapenemases and most of their gene variants (Table S2). The               desired range from 90% to 110%. BlaKPC (88%) and blaOXA-23 (89%)
aim was to cover at least 95% of the carbapenemases detected by the              were almost in this range. Only blaVIM-1 and blaIMP-8/24 showed poorer
NRL during the last years in Germany, which presented to be a                    efficiencies of 80% and 77% respectively (Fig. 1).
manageable threshold for our PCR design. The design of primers and
probes was based on the phylogeny of the different carbapenemase                 3.3. Validation of the automated assay
variants. All currently known blaKPC and blaNDM variants were detect­
able. Due to high sequence variations, the blaVIM-set covered all variants           The assay was validated as a fully automated PCR with the first
except blaVIM-7 (Fig. S1). BlaIMP and blaOXA type carbapenemase genes            collection of isolates from the NRL (n = 151 isolates). The collection
showed a high degree of sequence diversity, which is why not all vari­           contained 58 carbapenemase positive strains, 12 isolates with ESBL or
ants are detectable (Table 1, Fig. S2).                                          AmpC resistance genes and 81 beta-lactamase-negative isolates. All 58
                                                                                 carbapenemases were identified correctly. The oligonucleotides were
3.2. Assay performance                                                           specific for the particular carbapenemase and their tested variants, as
                                                                                 neither amplification of ESBLs (blaGES, blaSHV) and AmpC beta-
    The assay performance was determined on the BD MAX™ system,                  lactamases (blaDHA, blaCMY) nor of beta-lactamase-negative bacteria
using the PCR-only mode. One representative target of each carbapen­             could be observed, leading to a sensitivity and specificity of 100%
emase was selected to define the LOD (blaNDM-1/− 6, blaKPC-3, blaVIM-1,          (Table 2).
blaIMP-8/− 24, blaOXA-23, blaOXA-72, blaOXA-58, blaOXA-48). PCR was done             For further validation the second set of collected isolates from the

                                                                             3
K. Probst et al.                                                                                                       Journal of Microbiological Methods 185 (2021) 106224

Fig. 1. Determination of the limit of detection and PCR efficiencies. a: standard curve for MM1; the detection limit for blaNDM and blaVIM is 3.2 copies, for blaKPC and
blaIMP 32 copies. PCR efficiency in MM1 is 98% for blaNDM, 88% for blaKPC, 80% for blaVIM and 77% for blaIMP. b: standard curve for MM2; the detection limit for the
different blaOXA variants is 3.2 copies. PCR efficiencies are ranging from 89% to 99% in MM2.

                                                                                          identified correctly as well as blaNDM-5 and blaVIM-28 positive bacteria.
Table 2
                                                                                          Additionally, isolates that harbor two different carbapenemases in
Consecutive collection of carbapenemase-positive and -negative isolates, used
                                                                                          combination could be detected by the multiplex PCR (blaKPC-2 + blaVIM-
for the validation of the assay. BlaGES, blaCMY, blaDHA and blaSHV were considered
as “negative”.                                                                            26, blaNDM-1/6 + blaOXA-48, blaOXA-23 + blaOXA-58, blaNDM-5 + blaOXA-181,
                                                                                          Table S3).
  Beta-            Bacterial          Number      BD        Concordant   Discordant
  lactamase        species               of      MAX™
  NRL result                          isolates   result                                   3.4. Carbapenemase detection directly from rectal swabs
  blaVIM-1         K.oxytoca            13        blaVIM       13            0
                   E.cloacae                                                                  Apart from using the automated PCR for bacterial isolates, direct
                   K.pneumoniae                                                           detection from rectal swabs, e.g. during screening procedures, might be
                   E.coli                                                                 envisioned. For this purpose, 66 rectal swabs from Heidelberg hospital
  blaVIM-2         P.aeruginosa         10       blaVIM        10            0
                                                                                          patients were examined. Using the multiplex SYBR green real-time PCR
  blaKPC-2         K.pneumoniae          4       blaKPC         4            0
  blaKPC-3         K.pneumoniae          4       blaKPC         4            0            (Hofko et al., 2014) the carbapenemase type of each isolate, analyzed
  blaNDM-1/-6      E.coli, K.            5       blaNDM         5            0            during routine procedure, was determined. Concordant results were
                   pneumoniae,                                                            achieved in 77.3% (51/66 swabs), discordant results in 12.1% (8/66
                   A.baumannii
                                                                                          swabs) (Table 3). Using the direct PCR, six samples were false negative
  blaNDM-1/-       A.baumannii           1       blaNDM,        1            0
                                                                                          (3× blaVIM-1, 2× blaKPC, 1× blaIMP-1), and for one sample (blaNDM) the
    6,                                           blaOXA-
    blaOXA-                                       23-like
                                                                                          direct PCR failed due to inhibition, since no signal for 16S was detect­
    23                                                                                    able. In one case different carbapenemase types were detected
  blaOXA-23        A.baumannii          12       blaOXA-       12            0            (Table S4). The colony counts on the chromID® ESBL plates of these six
                                                  23-like
                                                                                          false-negative samples were all in the range of 1–5 colonies, which re­
  blaOXA-48        K.pneumoniae          9       blaOXA-        9            0
                   E.coli                         48-like
                                                                                          sembles a bacterial load of 100 CFU/ml-500 CFU/ml in the rectal swab.
  blaGES-1         P.aeruginosa          5       negative       5            0            In 7 cases, additional carbapenemase genes were detected in the direct
  blaCMY-2         E.coli                3       negative       3            0            PCR samples that were not identified during routine diagnostics (10.6%,
                   P.mirabilis
                                                                                          Table 3, Table S4).
  blaCMY-2/22      P.mirabilis           2       negative       2            0
  blaSHV-12        E.cloacae             1       negative       1            0
  blaDHA           E.coli                1       negative       1            0            4. Discussion
  Negative         P.aeruginosa,        81       negative      80            1a
                   E.cloacae, K.                                                              The aim of this study was to design a PCR-based assay for detecting
                   pneumoniae,
                   E.aerogenes,
                                                                                          eight common carbapenemases and their variants. This resulted in the
                   E.coli, A.                                                             design of two multiplex PCRs that could be run in a fully-automated
                   baumannii, P.                                                          manner to facilitate detection of carbapenemases from suspicious bac­
                   mirabilis, M.                                                          terial isolates or directly from rectal swabs.
                   morganii, A.
                                                                                              Multiple PCRs for the detection of carbapenemases have been pub­
                   pittii, A.junii,
                   X.maltophila,                                                          lished and are commercialized. However, a drawback of existing com­
                   C.freundii                                                             mercial assays is the fixed composition of included targets, whereas a
 a                                                                                        drawback of in-house assays is their greater demand for molecular
    The detection of 16S positive control failed due to inhibition. There was no
detectable signal for a carbapenemase gene.
                                                                                          expertise. So far, few carbapenemase assays have been reported for the
                                                                                          BD MAX™ (Hofko et al., 2014; Antonelli et al., 2016; Teo et al., 2016;
                                                                                          Teo et al., 2018). With the recent improvement allowing the use of two
NRL was tested, containing less common carbapenemase variants and
                                                                                          master mixes for one extraction the number of possible targets
combinations. Different blaOXA (blaOXA-58, blaOXA-72, blaOXA-204, blaOXA-
                                                                                          increased. The available assays cover almost all known variants of the
162, blaOXA-181, blaOXA-232, blaOXA-244) and blaIMP (blaIMP-1, blaIMP-13,
                                                                                          “big 4” carbapenemases, but especially for blaIMP and blaOXA some var­
blaIMP-16, blaIMP-7, blaIMP-14, blaIMP-15, blaIMP-4/28, blaIMP-2/19) types were
                                                                                          iants occurring in Germany in recent years (Koch-Institut, 2015; Koch-

                                                                                      4
K. Probst et al.                                                                                                          Journal of Microbiological Methods 185 (2021) 106224

Table 3
Carbapenemase detection directly from rectal swabs compared to differentiated isolates. In 77.3% (51/66 swabs) the carbapenemase genes were found in rectal swabs
and in the differentiated isolate. Discordant results were obtained in 12.1% (8/66 swabs), since the carbapenemase genes could not be identified in the direct PCR-
sample but were detected in isolates from selective agar or a different carbapenemase gene was detected. By direct PCR additional carbapenemase genes were detected
from rectal swabs (7/66, 10.6%), which were not found in routine diagnostics.
  Carbapenemases detected in cultured isolate from   Number of rectal                                  Results of direct PCR from rectal swabs
  rectal swab by SYBR green PCR                         swabs
                                                                        Concordant       Detection of new carbapenemase    Detection of different                Discordant
                                                                                         by direct PCR                     carbapenemase by direct PCR

  blaOXA-48-like                                           30              29                          1                                    0                        0
  blaOXA-23-like                                            4               3                          0                                    1                        0
  blaKPC                                                    9               5                          2                                    0                        2
  blaNDM                                                   11               7                          3                                    0                        1
  blaNDM, blaOXA-48-like                                    2               1                          1                                    0                        0
  blaVIM-1                                                  6               3                          0                                    0                        3
  blaVIM-2                                                  3               3                          0                                    0                        0
  blaIMP-1                                                  1               0                          0                                    0                        1

Institut, 2016; Pfennigwerth, 2017; Pfennigwerth, 2018) cannot be                        used copy numbers of bacterial genomes instead of CFU for quantifica­
detected. With the multiplex PCR developed by Teo et al. in 2016 and                     tion. However, we can estimate that 1 copy/μl equals approximately 1
2018 (Teo et al., 2016, Teo et al., 2018) blaIMP-7, blaIMP-13, blaIMP-14,                CFU/μl (1000 CFU/ml). Thus, dependent on the target our LOD seems to
blaIMP-15 and blaIMP-28 cannot be identified (Teo et al., 2016, Teo et al.,              be in line with the assays described by Antonelli et al. in 2016 with a
2018). The BD MAX Check-Points CPO assay is failing to detect blaIMP-14,                 LOD ranging from 73 CFU/ml to 90,000 CFU/ml (Antonelli et al., 2016),
blaIMP-15, blaIMP-16, blaIMP-18, blaIMP-22, blaIMP-23 and blaIMP-28 (Antonelli           and Lund et al. in 2018 with a LOD between 100 CFU/ml and 1,000
et al., 2016). The Cepheid Xpert Carba-R assay is designed for detecting                 CFU/ml (Lund et al., 2018), but might be less sensitive than the assays
blaNDM, blaKPC, blaVIM, blaIMP-1 and blaOXA-48-like, but blaOXA-23-like, blaOXA-         reported by Teo et al., 2016 (Teo et al., 2016) and Teo et al., 2018 (Teo
58-like and blaOXA-40/24-like are not included in the panel. Additionally,               et al., 2018) with LODs of 100 CFU/ml.
blaIMP-4, blaIMP-6, blaIMP-10 and blaIMP-11 are detectable with this assay                   The results encouraged us to use the assay for direct testing from
but blaIMP-7, blaIMP-13 and blaIMP-14 cannot be identified (Traczewski                   rectal swabs. Such a procedure reduces the time to results, since an
et al., 2018). These blaIMP-variants as well as blaOXA-40/24-like and blaOXA-            overnight culture is not necessary. Carbapenemase genes were found in
58-like previously not included but frequently found in A. baumannii                     59/66 rectal swabs (89.4%) by direct-PCR as compared to detection by
(Hofko et al., 2014; Antonelli et al., 2016; Teo et al., 2016; Teo et al.,               culture screen on chromID® ESBL plates (Table 3, Table S4). Other as­
2018) can be detected with our new designed assay (Table 1). However                     says designed for testing directly from rectal swabs showed higher
the Cepheid Xpert Carba-R assay, the BD MAX Check-Points CPO assay                       sensitivities ranging from 93.1% to 100% (Antonelli et al., 2016) and
and the assay designed by Teo et al. in 2016 (Teo et al., 2016) are able to              97.5%, respectively (Saliba et al., 2019). When testing directly from
detect blaIMP-4, which is not included in our assay (Antonelli et al., 2016;             rectal swabs, six samples gave false-negative results, in one case the PCR
Traczewski et al., 2018).                                                                failed due to inhibition. Other published assays also reported that some
     As the multiplex PCR was designed for the BD MAX™ system, which                     carbapenem resistant isolates can be overseen (limited sensitivity in
is beneficial for large laboratory use, other tests for the identification of            direct testing) (Antonelli et al., 2016; Lund et al., 2018; Saliba et al.,
carbapenemase-positive bacteria might be more suitable for locations                     2019). One possible reason could be the quantity of bacteria in the
with less carbapenemase prevalence and smaller diagnostic microbio­                      primary sample. This can be tested by counting the colonies on chro­
logical laboratories, like the carbapenem inactivation method (van der                   mID® ESBL plates. Typically, only one to five colonies grew on one plate
Zwaluw et al., 2015), CARBA-5 (Hopkins et al., 2018) or RESIST-4                         (resembling 100–500 CFU/ml), indicating that the amount of bacteria
(Cointe et al., 2018).                                                                   represents a detection limit for direct PCR application. However, in one
     Based on the data of the German NRL in the years 2014–2017 (Koch-                   case a blaKPC isolate with four colonies could be detected with a Ct value
Institut, 2015; Koch-Institut, 2016; Pfennigwerth, 2017; Pfennigwerth,                   of 26.4. Besides the number of bacteria, the number of plasmids, which
2018), 98.2% of the carbapenemases in Enterobacteriaceae, 94.0% in                       usually encode the resistance gene, could play a role in the detection
P. aeruginosa and 96.1% in A. baumannii are theoretically detectable                     limit. BlaKPC was described on low-copy number plasmids like IncF
with this assay, which leads to an overall theoretical sensitivity of                    plasmids as well as on high-copy number Col-like plasmids (Chavda
97.1%. Compared to the detection of carbapenemases in Enterobacteri­                     et al., 2015; Stoesser et al., 2017). In this case, perhaps the number of
aceae and A. baumannii the theoretical coverage of carbapenemases in                     blaKPC carrying plasmids could compensate the low bacterial load in the
P. aeruginosa was slightly lower (94.0%). Especially blaIMI and blaGIM                   rectal swab. Compared to other assays with efficiencies between 95%
carbapenemases occurred more frequently in P. aeruginosa during the                      and 105.2% (Teo et al., 2016; Lund et al., 2018), the amplification of
last years (Koch-Institut, 2015, Koch-Institut, 2016, Pfennigwerth, 2017,                blaVIM and blaIMP showed poorer efficiencies (80%, 77%) in our assay,
Pfennigwerth, 2018), which are not included in our assay.                                which could be another reason for the false-negative results (3× blaVIM-1,
     The assay was validated with a consecutive collection of 151 well                   1× blaIMP-1) obtained from rectal swabs, using the direct PCR. The ef­
characterized isolates from the NRL leading to a sensitivity and speci­                  ficiencies for blaNDM, blaKPC and blaOXA, however, are in line with
ficity of 100% by testing from bacterial colony, which is comparable to                  available assays.
other assays (Hofko et al., 2014; Teo et al., 2016; Lund et al., 2018; Teo                   However, our assay identified different or new carbapenemase genes
et al., 2018; Mentasti et al., 2019).                                                    that were not detected by the routine procedure (8/66 swabs). These
     Our assay showed a good LOD in the PCR-only mode, since 1 copy/μl                   carbapenemase-positive bacteria perhaps were not viable/cultivable
for blaNDM, blaVIM and blaOXA and 10 copies/μl for blaKPC and blaIMP were                anymore due to antibiotic treatment or unculturability of the carrier,
detectable, which equals 3.2 copies and 32 copies per PCR, respectively                  since mixed populations can be present in the rectal swab. Another
(Fig. 1). The detection limit of other assays was 73 CFU/ml to 900,000                   possibility is that the resistance gene is present but the expression level is
CFU/ml (Antonelli et al., 2016), 100 CFU/ml (Teo et al., 2016; Teo et al.,               low or the gene is not expressed at all, which leads to phenotypical
2018) and 100 CFU/ml to 1,000 CFU/ml, respectively (Lund et al.,                         sensitivity. For culture-negative rectal swabs, isolates might have grown
2018). It is difficult to compare our LOD with those of other assays, as we              on selective agar with a further enrichment culture. Identification of

                                                                                     5
K. Probst et al.                                                                                                        Journal of Microbiological Methods 185 (2021) 106224

additional carbapenemases is in line with reported results from other              analysis and interpretation of data: K.P., S.B., A.H.D.; drafting the article
validation of molecular assays such as Check-Direct CPE real-time PCR              or revising it critically for important intellectual content: K.P., S.B., M.
(Antonelli et al., 2016) and GeneXpert® (Saliba et al., 2019). The data            B., K.H., A.H.D.
indicate that neither molecular nor culture screening detect all present
carbapenem resistant bacteria.                                                     Declaration of interests
    The following limitations could be observed: To determine the
relative LOD genomic copy numbers of the bacteria were used. Thus,                     The authors declare that they have no known competing financial
depending on the copy numbers of the carbapenemase encoding plasmid                interests or personal relationships that could have appeared to influence
the LOD may vary between different isolates (Chavda et al., 2015;                  the work reported in this paper.
Stoesser et al., 2017; Shen et al., 2020). Whether the carbapenemase
detection from direct PCR-samples is more sensitive than culture-based             Acknowledgement
methods cannot be determined from our test setting, since only
carbapenemase-positive rectal swabs were used for analyzing the assay                 We thank Martin Kaase, National Reference Laboratory for
performance, and is thus subject to further investigation. For identifi­           Multidrug-Resistant Gram-Negative Bacteria, Bochum, Germany for
cation of carbapenemase-positive bacteria chromID® ESBL plates were                providing samples as published before (Hofko et al., 2014).
used as a selective media. The detection limit of this agar is dependent
on the bacterial species, the present carbapenemase and also the mini­             Appendix A. Supplementary data
mum inhibitory concentration to the antibiotics (Nordmann et al., 2012;
Vrioni et al., 2012; Göttig et al., 2020). Especially OXA-48-producing               Supplementary data to this article can be found online at https://doi.
bacteria that do not co-express an ESBL are inhibited in growth                    org/10.1016/j.mimet.2021.106224.
(Göttig et al., 2020), which might also be an explanation for detecting
more carbapenemases by direct PCR. Another drawback of our assay is
                                                                                   References
the limited number of isolates and clinical samples with different
carbapenemase variants that could be experimentally tested. Particu­               Albiger, B., Glasner, C., Struelens, M.J., Grundmann, H., Monnet, D.L., t. E. S. o. C.-P. E.
larly blaVIM-4 and blaVIM-11 were detected more frequently by the NRL                   w. Group, 2015. Carbapenemase-producing Enterobacteriaceae in Europe:
                                                                                        assessment by national experts from 38 countries, May 2015. Eurosurveillance 20
during the last years (Koch-Institut, 2015; Koch-Institut, 2016; Pfen­
                                                                                        (45), 30062.
nigwerth, 2017; Pfennigwerth, 2018; Pfennigwerth, 2019; Pfennig­                   Antonelli, A., Arena, F., Giani, T., Colavecchio, O.L., Valeva, S.V., Paule, S., Boleij, P.,
werth, 2020). Other blaVIM, blaNDM, blaKPC, blaIMP and blaOXA variants,                 Rossolini, G.M., 2016. Performance of the BD MAX instrument with Check-Direct
not present in our collection, occurred in less than 1% of the analyzed                 CPE real-time PCR for the detection of carbapenemase genes from rectal swabs, in a
                                                                                        setting with endemic dissemination of carbapenemase-producing
carbapenemase positive bacteria by the NRL from 2014 to 2019 (Koch-                     Enterobacteriaceae. Diagn. Microbiol. Infect. Dis. 86 (1), 30–34.
Institut, 2015, Koch-Institut, 2016, Pfennigwerth, 2017, Pfennigwerth,             Chavda, K.D., Chen, L., Jacobs, M.R., Rojtman, A.D., Bonomo, R.A., Kreiswirth, B.N.,
2018, Pfennigwerth, 2019, Pfennigwerth, 2020). Nonetheless if new                       2015. Complete sequence of a bla(KPC)-harboring cointegrate plasmid isolated from
                                                                                        Escherichia coli. Antimicrob. Agents Chemother. 59 (5), 2956–2959.
carbapenemase variants appear during routine diagnostics, they can be              Cointe, A., Bonacorsi, S., Truong, J., Hobson, C., Doit, C., Monjault, A., Bidet, P.,
validated by PCR. Additionally, rare carbapenemases like blaGIM or                      Birgy, A., 2018. Detection of carbapenemase-producing enterobacteriaceae in
blaGES cannot be detected, which could lead to false negative results.                  positive blood culture using an immunochromatographic RESIST-4 O.K.N.V. assay.
                                                                                        Antimicrob. Agents Chemother. 62 (12).
This limitation could be resolved, however, by adding more probes to               Cuzon, G., Naas, T., Bogaerts, P., Glupczynski, Y., Nordmann, P., 2012. Evaluation of a
the assay. Furthermore, our assay does not identify the bacterial species,              DNA microarray for the rapid detection of extended-spectrum beta-lactamases (TEM,
which would be helpful for antibiotic treatment.                                        SHV and CTX-M), plasmid-mediated cephalosporinases (CMY-2-like, DHA, FOX,
                                                                                        ACC-1, ACT/MIR and CMY-1-like/MOX) and carbapenemases (KPC, OXA-48, VIM,
                                                                                        IMP and NDM). J. Antimicrob. Chemother. 67 (8), 1865–1869.
5. Conclusion                                                                      Dallenne, C., Da Costa, A., Decre, D., Favier, C., Arlet, G., 2010. Development of a set of
                                                                                        multiplex PCR assays for the detection of genes encoding important beta-lactamases
                                                                                        in Enterobacteriaceae. J. Antimicrob. Chemother. 65 (3), 490–495.
    The developed assay for the BD MAX™ system provides a time
                                                                                   Doumith, M., Ellington, M.J., Livermore, D.M., Woodford, N., 2009. Molecular
effective method to detect carbapenemases that frequently occur in                      mechanisms disrupting porin expression in ertapenem-resistant Klebsiella and
Germany from bacterial colonies in less than 3 h. The automatization of                 Enterobacter spp. clinical isolates from the UK. J. Antimicrob. Chemother. 63 (4),
the assay allows an easy performance of DNA extraction and real-time                    659–667.
                                                                                   El Amin, N., Giske, C.G., Jalal, S., Keijser, B., Kronvall, G., Wretlind, B., 2005.
PCR without requiring in-depth experience. The assay is rapid and re­                   Carbapenem resistance mechanisms in Pseudomonas aeruginosa: alterations of porin
quires little hands-on time. Due to the usage of the BD MAX™ in an open                 OprD and efflux proteins do not fully explain resistance patterns observed in clinical
mode, the user-developed protocol could be adapted in case new vari­                    isolates. APMIS 113 (3), 187–196.
                                                                                   Göttig, S., Walker, S.V., Saleh, A., Koroska, F., Sommer, J., Stelzer, Y., Steinmann, J.,
ants appear or epidemiology changes. With two multiplex PCRs eight                      Hamprecht, A., 2020. Comparison of nine different selective agars for the detection
important carbapenemases and most of their known variants occurring                     of carbapenemase-producing Enterobacterales (CPE). Eur. J. Clin. Microbiol. Infect.
in Enterobacteriaceae and the non-fermenters P. aeruginosa and                          Dis. 39 (5), 923–927.
                                                                                   Hofko, M., Mischnik, A., Kaase, M., Zimmermann, S., Dalpke, A.H., 2014. Detection of
A. baumannii could successfully be detected. The assay showed high                      carbapenemases by real-time PCR and melt curve analysis on the BD Max system.
sensitivity and specificity, and might also be adapted to initial screening             J. Clin. Microbiol. 52 (5), 1701–1704.
of patients possibly colonized with carbapenemase harboring bacteria.              Hopkins, K.L., Meunier, D., Naas, T., Volland, H., Woodford, N., 2018. Evaluation of the
                                                                                        NG-Test CARBA 5 multiplex immunochromatographic assay for the detection of
The shorter time to results can avoid isolation of patients and will help to            KPC, OXA-48-like, NDM, VIM and IMP carbapenemases. J. Antimicrob. Chemother.
prevent spread of antimicrobial resistances and possible outbreaks by                   73 (12), 3523–3526.
allowing timely initiation of infection control measures.                          Kaase, M., 2012. Carbapenemases in gram-negative bacteria. Current data and trends of
                                                                                        resistance resulting from the work of national reference centres.
                                                                                        Bundesgesundheitsbl. Gesundheitsforsch. Gesundheitsschutz 55 (11− 12),
Funding                                                                                 1401–1404.
                                                                                   Koch-Institut, R., 2015. Bericht des Nationalen Referenzzentrums (NRZ) für
                                                                                        gramnegative Krankenhauserreger. Robert Koch-Institut, Epidemiologie und
    This work was supported by a grant from the German Federal Min­
                                                                                        Gesundheitsberichterstattung.
istry of Education and Research (BMBF) (16GW0163).                                 Koch-Institut, R., 2016. Bericht des Nationalen Referenzzentrums (NRZ) für
                                                                                        gramnegative Krankenhauserreger. Robert Koch-Institut, Epidemiologie und
Authors’ contributions                                                                  Gesundheitsberichterstattung.
                                                                                   Lund, M., Petersen, M.B., Jorgensen, A.L., Paulmann, D., Wang, M., 2018. Rapid real-
                                                                                        time PCR for the detection of IMP, NDM, VIM, KPC and OXA-48 carbapenemase
    Conception and design of the study: A.H.D., K.H.; acquisition,                      genes in isolates and spiked stool samples. Diagn. Microbiol. Infect. Dis. 92 (1), 8–12.

                                                                               6
K. Probst et al.                                                                                                                           Journal of Microbiological Methods 185 (2021) 106224

Martin-Platero, A.M., Peralta-Sanchez, J.M., Soler, J.J., Martinez-Bueno, M., 2010.                    Teo, J.W., Octavia, S., Cheng, J.W., Lin, R.T., 2018. Evaluation of an in-house developed
    Chelex-based DNA isolation procedure for the identification of microbial                               multiplex real-time PCR for the detection of IMP, OXA-23, GES carbapenemases and
    communities of eggshell surfaces. Anal. Biochem. 397 (2), 253–255.                                     the transmissible colistin-resistant mcr gene on the BD MAX open system. Diagn.
Mentasti, M., Prime, K., Sands, K., Khan, S., Wootton, M., 2019. Rapid Detection of IMP,                   Microbiol. Infect. Dis. 90 (1), 67–69.
    NDM, VIM, KPC and OXA-48-like carbapenemases from Enterobacteriales and Gram-                      Traczewski, M.M., Carretto, E., Canton, R., Moore, N.M., 2018. Multicenter evaluation of
    negative non-fermenter bacteria by real-time PCR and melt-curve analysis. Eur. J.                      the Xpert Carba-R assay for detection of carbapenemase genes in gram-negative
    Clin. Microbiol. Infect. Dis. 38 (11), 2029–2036.                                                      isolates. J. Clin. Microbiol. 56 (8) e00272–00218.
Naas, T., Nordmann, P., 1994. Analysis of a carbapenem-hydrolyzing class A beta-                       van der Zwaluw, K., de Haan, A., Pluister, G.N., Bootsma, H.J., de Neeling, A.J.,
    lactamase from Enterobacter cloacae and of its LysR-type regulatory protein. Proc.                     Schouls, L.M., 2015. The carbapenem inactivation method (CIM), a simple and low-
    Natl. Acad. Sci. U. S. A. 91 (16), 7693–7697.                                                          cost alternative for the Carba NP test to assess phenotypic carbapenemase activity in
Naas, T., Cuzon, G., Bogaerts, P., Glupczynski, Y., Nordmann, P., 2011. Evaluation of a                    gram-negative rods. PLoS One 10 (3), e0123690.
    DNA microarray (Check-MDR CT102) for rapid detection of TEM, SHV, and CTX-M                        Vrioni, G., Daniil, I., Voulgari, E., Ranellou, K., Koumaki, V., Ghirardi, S., Kimouli, M.,
    extended-spectrum beta-lactamases and of KPC, OXA-48, VIM, IMP, and NDM-1                              Zambardi, G., Tsakris, A., 2012. Comparative evaluation of a prototype chromogenic
    carbapenemases. J. Clin. Microbiol. 49 (4), 1608–1613.                                                 medium (ChromID CARBA) for detecting carbapenemase-producing
Nadkarni, M.A., Martin, F.E., Jacques, N.A., Hunter, N., 2002. Determination of bacterial                  Enterobacteriaceae in surveillance rectal swabs. J. Clin. Microbiol. 50 (6),
    load by real-time PCR using a broad-range (universal) probe and primers set.                           1841–1846.
    Microbiology 148 (Pt 1), 257–266.
Nordmann, P., Girlich, D., Poirel, L., 2012. Detection of Carbapenemase producers in
    Enterobacteriaceae by use of a novel screening medium. J. Clin.
    Microbiol. 50 (8), 2761–2766.
                                                                                                       CFU: colony forming unit
Pfennigwerth, N., 2017. Bericht des Nationalen Referenzzentrums (NRZ) für
                                                                                                       Ct: cycle threshold
    gramnegative Krankenhauserreger. Robert Koch-Institut, Epidemiologie und
                                                                                                       Cy5: cyanine5
    Gesundheitsberichterstattung.
                                                                                                       Cy5.5: cyanine5.5
Pfennigwerth, N., 2018. Bericht des Nationalen Referenzzentrums (NRZ) für
                                                                                                       ESBL: extended-spectrum β-lactamase
    gramnegative Krankenhauserreger. Robert Koch-Institut, Epidemiologie und
                                                                                                       FAM: 6-carboxyfluorescein
    Gesundheitsberichterstattung.
                                                                                                       GES: Guiana extended-spectrum β-lactamase
Pfennigwerth, N., 2019. Bericht des Nationalen Referenzzentrums für gramnegative
                                                                                                       GIM: German imipenemase
    Krankenhauserreger. Epidemiol. Bull. 31, 289–294.
                                                                                                       IMP: imipenem-hydrolyzing β-lactamase
Pfennigwerth, N., 2020. Bericht des Nationalen Referenzzentrums für gramnegative
                                                                                                       JOE: 6-Carboxy-4′ ,5’-Dichloro-2′ ,7’-Dimethoxyfluorescein, Succinimidyl Ester
    Krankenhauserreger. Epidemiol. Bull. 26, 3–10.
                                                                                                       KPC: Klebsiella pneumoniae carbapenemase
Queenan, A.M., Bush, K., 2007. Carbapenemases: the versatile beta-lactamases. Clin.
                                                                                                       LOD: limit of detection
    Microbiol. Rev. 20 (3), 440–458 table of contents.
                                                                                                       MALDI-TOF: Matrix-assisted laser desorption/ionization-time of flight
Saliba, R., Neulier, C., Seytre, D., Fiacre, A., Faibis, F., Leduc, P., Amara, M., Jaureguy, F.,
                                                                                                       MUSCLE: Multiple Sequence Comparison by Log-Expectation
    Carbonnelle, E., Zahar, J.R., Marty, L., 2019. Can real-time polymerase chain
                                                                                                       NCBI: National Center for Biotechnology Information
    reaction allow a faster recovery of hospital activity in cases of an incidental
                                                                                                       NDM: New Delhi metallo-β-lactamase
    discovery of carbapenemase-producing Enterobacteriaceae and vancomycin-
                                                                                                       NRL: National Reference Laboratory
    resistant Enterococci carriers? J. Hosp. Infect. 103 (2), 115–120.
                                                                                                       OXA-23: oxacillinase-23
Shen, Z., Zhang, H., Gao, Q., Qin, J., Zhang, C., Zhu, J., Li, M., 2020. Increased plasmid
                                                                                                       OXA-40/24: oxacillinase-40/24
    copy number contributes to the elevated carbapenem resistance in OXA-232-
                                                                                                       OXA-48: oxacillinase-48
    producing Klebsiella pneumoniae. Microb. Drug Resist. 26 (6), 561–568.
                                                                                                       OXA-58: oxacillinase-58
Stoesser, N., Sheppard, A.E., Peirano, G., Anson, L.W., Pankhurst, L., Sebra, R., Phan, H.
                                                                                                       PCR: polymerase chain reaction
    T.T., Kasarskis, A., Mathers, A.J., Peto, T.E.A., Bradford, P., Motyl, M.R., Walker, A.
                                                                                                       UPGMA: Unweighted Pair Group Method with Arithmetic mean
    S., Crook, D.W., Pitout, J.D., 2017. Genomic epidemiology of global Klebsiella
                                                                                                       VIM: Verona integron-encoded metallo-β-lactamase
    pneumoniae carbapenemase (KPC)-producing Escherichia coli. Sci. Rep. 7 (1), 5917.
Teo, J.W., La, M.V., Lin, R.T., 2016. Development and evaluation of a multiplex real-time
    PCR for the detection of IMP, VIM, and OXA-23 carbapenemase gene families on the
    BD MAX open system. Diagn. Microbiol. Infect. Dis. 86 (4), 358–361.

                                                                                                   7
You can also read